2014-07-23 16:35:06 -0600 | commented question | Parallel Random Trees/Forest also keep in mind that there are different ways to parallel. there is shared memory, global file system and mapreduce-type. |
2014-07-23 16:34:11 -0600 | answered a question | Parallel Random Trees/Forest A naive solution would be to run random forests in parallel and then to average the predictions weighted by probability. |
2014-07-23 16:31:44 -0600 | commented question | Parallel Random Trees/Forest java's spark (based on hadoop) has a parallel tree-building algorithm. there is also a paper by google on mapreduce and rf http://static.googleusercontent.com/media/research.google.com/en/us/pubs/archive/36296.pdf. i'm also trying to figure out how to use open cv in parallel |
2014-07-23 16:29:32 -0600 | received badge | ● Supporter (source) |
2014-02-20 13:33:25 -0600 | received badge | ● Scholar (source) |
2014-02-19 15:33:25 -0600 | commented answer | Trivial random forest with OpenCV doesn't work and isn't the same as sklearn i increased the training data to 3 points and it worked. might be better to put the comment as an answer. |
2014-02-17 17:35:50 -0600 | commented answer | Trivial random forest with OpenCV doesn't work and isn't the same as sklearn I tried that initially. I forgot to change it back. It should be ROW. I'll edit the question. thanks. |
2014-02-17 12:55:08 -0600 | received badge | ● Editor (source) |
2014-02-17 12:54:24 -0600 | asked a question | Trivial random forest with OpenCV doesn't work and isn't the same as sklearn I'm trying to get the simplest example of random forest to work. The training data is 2 points {0,0} with a label 0 and {1,1} with a label 1. The sample to predict is {2,2}. OpenCV returns 0 rather than 1. Here is the OpenCV code in C++ ( cmake file: running: To compare, here is a python's sklearn code ( running: Update (1) I tried different combination of placement of data and also changed the line to I still get just 0. |